The new agreement takes into account the development and changes of the banking industry, especially new businesses such as mixed banking operations and asset securitization. The impact of the development of new products is based on the 1999 Agreement after absorbing opinions from many parties. The revised results of the original framework announced in June. This shows that the new agreement has extensively considered the current development status of the banking industry and has certain coverage. Overall, the new agreement embodies the following main features:

1. Breaking through the limitations of the traditional banking industry. The new agreement itself takes into account the consolidation of different institutions under holding companies. In terms of products, it covers capital requirements for securitized assets and securities held by banks. At the same time, the Basel Committee has also begun to promote cooperation with insurance industry regulators to further Promote the development of new rules. The new agreement promotes the scope of application of the classic minimum capital ratio in terms of institutions and business types, which establishes an important policy basis for the formation of merger supervision of the financial industry in an all-round development environment of the banking industry.

2. More flexible and dynamic rules. The new agreement allows banks to implement internal rating methods, giving the new regulatory rules a certain degree of flexibility and conducive to absorbing various advanced experiences in risk management of modern large banks. In order to encourage the update of the capital requirement method established in Pillar 1, the new agreement encourages banks to continuously improve risk assessment methods and continuously develop more sophisticated risk assessment systems. At the same time, banks are also encouraged to adopt advanced internal rating methods when they have sufficient data. The new agreement is conducive to promoting the advancement of risk management technology in the modern banking industry.

3. Pay attention to the combination of qualitative and quantitative aspects, and make the quantitative aspect more refined. The new agreement builds a new policy framework with three pillars and emphasizes the need for coordinated development of the three pillars. It is a combination of quantitative (capital calculation) and qualitative (requirements for regulatory processes, bank management systems and the use of market discipline rules). .

As we all know, quantitative calculation of capital is important, but due to the difficulty in obtaining data and the difficulty in measuring some risks, complete quantification cannot be achieved. Therefore, system construction and process control are very important supplements. The new rules also emphasize both quantitative and qualitative requirements for information disclosure. Compared with the 1988 policy framework, the new agreement's quantitative calculations are more refined. For example, in the internal rating method, risk estimation uses multiple variables such as the borrower's probability of default (PD), loan-specific loss given default (LGD), and exposure at default (EAD). The bank's risk aggregation takes into account the characteristics of various risks. Correlation introduces more complex non-linear relationships into bank risk measurement, which is undoubtedly more realistic. The new agreement will have extremely important implications for international banking regulation and the way many banks operate. First of all, it should be pointed out that the new agreement, which is characterized by three major elements (capital adequacy ratio, regulatory supervision and inspection, and market discipline), represents the development trend and direction of capital supervision. Practice has proved that capital adequacy ratio alone cannot guarantee the stability of a single bank or even the entire banking system. Since the advent of the Capital Accord in 1988, regulatory authorities in some countries have used these three means to strengthen capital supervision to varying degrees at the same time to achieve the goal of sound bank operations. However, organically combining the three major elements and fixing them in the form of regulatory regulations requires regulatory authorities to implement them seriously. This is undoubtedly an affirmation of successful regulatory experience and a major breakthrough in the field of capital regulation.

Different from the 1988 Capital Accord, the Basel Committee hoped from the beginning that the scope of application of the new agreement would not be limited to the G10 countries, although its focus is still on the country's "internationally active banks" (internationally active banks). The Basel Committee proposed that the basic principles of the New Capital Accord are generally applicable to all banks around the world and expected that many banks in non-Group 10 countries will use the standardized method to calculate minimum capital requirements. In addition, the Basel Committee also hopes that, over time, all the world's major banks will comply with the new agreement. Objectively speaking, once the new agreement comes out, participants in the international financial market are likely to use the new agreement to analyze the capital status of banks in various countries, and relevant international organizations will also regard the new agreement as a new international standard for bank supervision and assist Barcelona. The Al Commission promotes the new agreement globally and examines its implementation. Therefore, developing countries need to carefully study the impact of the new agreement.

Compared with the 1988 Capital Accord, the content of the New Capital Accord is broader and more complex.

This is because the new agreement strives to closely combine capital adequacy ratios with the main risks faced by banks, strives to reflect the latest changes in bank risk management and supervisory practices, and provides as much information as possible for banking industries and bank supervision systems with different levels of development. Item selection method. It should be said that the complexity of the banking regulatory system is entirely determined by the complexity of the banking system itself. Banks in the G10 countries will implement the new agreement within a specified time. In order to ensure their position in international competition, non-G10 countries will also strive to fully implement the new agreement within the specified time. Compared with developed countries, there is a large gap in the market development and regulatory level of developing countries, and the difficulty of implementing the new agreement cannot be underestimated. Here, it must also be pointed out that as far as the plan is concerned, the new agreement is first and foremost an agreement between the G10 countries and has not fully taken into account the national conditions of developing countries.

The New Basel Capital Accord proposes two methods for dealing with credit risk: the standardized method and the internal ratings method. The standardized approach is based on the 1988 Capital Accord and uses external rating agencies to determine risk weights, targeting banks with low complexity. The use of external rating agencies should be said to be more objective and better reflective of actual risk levels than the original classification method based on OECD countries. However, for the majority of developing countries, including China, to a considerable extent, the objective conditions for using this law do not exist. There are only a few domestic rating companies in developing countries, and it is difficult to meet internationally recognized standards; the number of banks and companies that have received ratings is limited; the cost of rating is high, and the results may not be objective and reliable. If the provisions of the standard method are applied rigidly, the rating of the vast majority of companies will be lower than BBB, and the risk weight will be 100%, or even 150% (for companies below BB-). Companies will have no incentive to participate in ratings because the risk weight of unrated companies is only 100%. Furthermore, taking this approach will naturally lead to a general increase in banks' capital levels due to increased risk weights and the introduction of capital requirements for operational risk.

Using the internal ratings-based approach to capital supervision is the core content of the new Capital Accord. This approach inherits the innovations of the 1996 Market Risk Supplement, which allows the use of its own internal measurement data to determine capital requirements. There are two forms of internal ratings-based approach, primary approach and advanced approach. The primary method only requires banks to calculate the borrower's default probability, and other risk factor values ??are determined by the regulatory authorities. Advanced rules allow banks to use multiple self-calculated risk element values. In order to promote the use of the internal ratings-based approach, the Basel Committee arranged a three-year transition period starting from 2004 for banks adopting this approach. First, it should be said that the new Basel Agreement has further improved the old Basel Agreement. It fully considers the various risks that banks may face; it has strong flexibility, mainly reflected in providing banks with multiple choices in the method of evaluating asset risks; in addition, the information disclosure requirements will also make banks more transparent to the public.

Second, the New Basel Agreement also has shortcomings.

The first is the issue of sovereign risk + although the status of country-specific standards has declined, it still plays a role in banks' asset selection, and its potential influence cannot be underestimated.

The second is the issue of risk weight. If the indicators are determined by the supervisory authority, it will be difficult to fully guarantee the objectivity, fairness and science of the authority's indicator selection. If it is left to the banks to decide on their own, such problems will also exist.

The third issue is the applicability of measurement methods: the new agreement encourages banks to use measurement methods based on internal ratings, but they have long-term operating records, and therefore have sufficient data and powerful technology to process these data efficiently. After all, there are only a few large banks with strong financial strength, and most banks still find it difficult to get rid of their reliance on external ratings and indicators recommended by the authorities.

Fourth, the main objects of supervision are commercial banks. However, under the general trend of financial internationalization, bank department stores continue to emerge, non-bank financial institutions and non-bank financial businesses continue to rise. In this regard, the role of the new agreement Space will be very limited. The IRB uses the same risk-weighted asset calculation methodology for sovereign, bank and corporate exposures.

This method relies on four aspects of data. One is the probability of default (PD), which is the possibility of the borrower defaulting on a specific period of time; the other is the loss given default (LGD), which is the risk when a default occurs. The degree of loss exposed; the third is Exposure at default (EAD), that is, for a certain loan commitment, the loan amount that may be withdrawn in the event of default; the fourth is Maturity (M), that is, a certain risk The remaining economic maturity of the exposure.

The corporate risk weight function specifies a specific capital requirement for each risk exposure after taking into account four parameters simultaneously. In addition, for loans to small and medium-sized enterprises defined as annual sales of less than 50 million euros, banks can adjust the company's IRB risk weight formula according to the size of the enterprise.

The main difference between the IRB advanced approach and the primary approach is reflected in the data requirements. The data required by the former are the bank's own estimates, while the data required by the latter are determined by the supervisory authority. The difference in this regard is shown in the table below:

Data IRB Junior Method IRB Advanced Method

Probability of Default (PD) Estimates provided by banks Estimates provided by banks

< p>Loss Given Default (LGD) Estimates provided by banks as regulatory indicators specified by the Committee

Exposure at Default (EAD) Estimates provided by banks as regulatory indicators specified by the Committee

Term ( M) Supervisory indicators specified by the Committee or

at the discretion of each country's supervisory authorities, estimates provided by banks (but excluding certain risk exposures) are allowed to be used Estimates provided by banks (but excluding certain risk exposures) )

The above table shows that for corporate, sovereign and interbank exposures, all banks using the IRB approach must provide internal estimates of default probabilities. In addition, banks that adopt the IRB advanced approach must provide internal estimates of LGD and EAD, while banks that adopt the IRB primary approach will use the indicators specified by the supervisory authorities in the third draft after taking into account the nature of the risk exposure. In general, banks that adopt the IRB advanced approach should provide estimates of the remaining maturities of the above types of risk exposures. However, it does not rule out that in individual cases, supervisory authorities may allow the use of fixed maturity assumptions. For banks adopting the IRB primary approach, national supervisory authorities can decide for themselves whether all banks in the country adopt the fixed maturity assumption specified in the third draft, or whether the banks provide their own estimates of remaining maturity.

Another important aspect of the IRB Law is the treatment of credit risk mitigation instruments (ie: mortgages, guarantees and credit derivatives). The IRB method itself, particularly the LGD parameters, provides sufficient flexibility for assessing the potential value of credit risk mitigation instrument technologies. Therefore, for banks adopting the IRB primary approach, the different LGD values ??specified by supervisors in the third draft reflect the existence of different categories of collateral. Banks that adopt the IRB advanced approach have greater flexibility when assessing the value of different categories of collateral. For transactions involving financial collateral, the IRB Act seeks to ensure that banks use recognized methods to assess risk as the value of the collateral changes. The third draft sets out a clear set of criteria for this. For retail risk exposures, only the IRB advanced approach can be used, and the IRB junior approach cannot be used. The main data of the IRB retail risk exposure formula are PD, LGD and EAD, which are entirely estimates provided by the bank. Compared with the IRB method of corporate risk exposure, there is no need to calculate a single risk exposure here, but it is necessary to calculate the estimated value of a basket of similar risk exposures.

Considering that various products included in retail risk exposures show different historical losses, retail risk exposures are divided into three major categories. The first is exposures secured by residential mortgage loans, the second is qualifying revolving retail exposures (QRRE), and the third is other non-revolving retail exposures, also known as other retail exposures. Generally speaking, the QRRE category includes various types of unsecured recurring retail exposures with specific loss characteristics, including various types of credit cards. All other non-housing consumer loans, including small business loans, are listed under other retail exposures. The third draft stipulates different risk weight formulas for these three types of businesses.

Specialized lending

Basel II subdivides wholesale loans that are different from other corporate loans and calls them collectively specialized loans. Professional loans refer to financing provided by a single project, the repayment of which is closely related to the operation of the corresponding asset pool or collateral. For all but one specialty loan, if banks can meet the minimum requirements for estimating relevant data, they can use the corporate loan IRB method to calculate the risk weights for such risk exposures. However, considering that there are still many difficulties in meeting these requirements in practice, the third draft also requires banks to subdivide such risk exposures into five levels. The third draft stipulates clear risk weights for each tranche.

For a specific category of specialist loans, 'high volatility commercial real estate' (HVCRE), IRB banks with the ability to estimate the required data will use a separate formula. Because these types of loans carry greater risk, this formula is more conservative than the typical corporate loan risk weight formula. For banks unable to estimate the required data, Cocoa breaks down HVCRE risk exposures into five categories. The third draft clearly stipulates the risk weights for each tranche. (Equity exposures)

IRB banks need to handle equity risk exposures separately. The third draft provides for two approaches. The first approach builds on the PD and LGD of corporate loans and requires banks to provide estimates of the underlying equity exposures. However, this approach specifies a LGD of 90% and imposes other restrictions, including a minimum risk weight of 100% in many cases. The second approach is designed to encourage banks to model the likelihood of a fall in the market value of equity holdings over a quarter. A simplified approach is also provided here which includes fixed risk weights for equity interests in listed and unlisted companies.